Bugzilla – Bug 229217
OpenSuse 10.2 Installation freezes at Starting udev...
Last modified: 2007-02-28 13:33:26 UTC
Have been using Suse 9.0 up to 10 without any such issue on another PC. Installed 10.1 on present machine without any installation issue. During installation of OpenSuse 10.2 right after choosing either text or GUI mode, the system freezes at Starting udev... I've waited up to 3 minutes on various attempts but system stalls at that process. My media is DVD x86_64. MD5 is OK. Media Check is OK via Yast. My PC is a Presario SR2030NX Suse 10.1 is the only OS on that machine.
Might be similar to 228195 since system also has SATA drive but user in bug 228195 seems to get past udev issue that I'm having.... At this point I don't have any logs to contribute since it's basically what's reported on screen. If there's a way to capture this let me know and I will gladly attach the result.
We are going to need some form of kernel log message in order to figure out what this is. Can you boot in the "safe" or "acpi-off" mode?
(In reply to comment #0) > Have been using Suse 9.0 up to 10 without any such issue on another PC. > Installed 10.1 on present machine without any installation issue. > > During installation of OpenSuse 10.2 right after choosing either text or GUI > mode, the system freezes at Starting udev... > I've waited up to 3 minutes on various attempts but system stalls at that > process. > My media is DVD x86_64. MD5 is OK. Media Check is OK via Yast. > My PC is a Presario SR2030NX Suse 10.1 is the only OS on that machine. > Hi Greg, Both Safe Mode and ACPI disabled work just fine, I've just tried that quickly this morning then stopped the install. When I do the regular install I hand at Starting udev... I can't even issue a ctrl-alt-F2 command to try to get something for you. Is there a particular time or other command that i should try?
This seems to be the initialization problem we keep on having with 10.2. Reassigning to ACPI expert.
I wonder whether this is also a duplicate of: https://bugzilla.novell.com/show_bug.cgi?id=232013 Do you have an nforce chipset? (what says: "grep -i nforce /var/log/messages" or "hwinfo |grep -i nforce") Can you check whether a 32-bit medium can boot, pls.
hwinfo | grep -i nforce <6>NFORCE-MCP51: IDE controller at PCI slot 0000:00:0d.0 <6>NFORCE-MCP51: chipset revision 161 <6>NFORCE-MCP51: not 100% native mode: will probe irqs later <6>NFORCE-MCP51: 0000:00:0d.0 (rev a1) UDMA133 controller <6>forcedeth.c: Reverse Engineered nForce ethernet driver. Version 0.49. Sorry I don't have time right now to test 32-bit Opensuse 10.2. I will see later in the weekend and get back to you. I don't think I will have problems though since my 10.1 was 32 bit.
It should also boot with: noapic nosmp Like that you have ACPI functionality. Is this a one processor core machine? Possibly only nolapic is enough?
Hi Thomas I'm getting the 32-bit DVD at the moment to test as you requested. Please explain the boot parameters noapic, nosmp, nolapic. I suppose nosmp is no multiple processor which would be correct since it's a one processor machine to answer one of your question. You can use the link below to check the motherboard specs to see if you need any other info. http://h10025.www1.hp.com/ewfrf/wc/document?docname=c00757470&lc=en&cc=tr&dlc=en&product=3245050&lang=en I will burn 32-bit as soon as it's finished and test it and the 64-bit with different iterations of the parameters you provided.
Hi Thomas, The 32-bit version works fine. With the 64-bit version and parameters you provided: The noapic boot option alone works. Adding the nosmp option with it also works. However, using only the nosmp option ( I wanted to check) makes the installation process freeze at driver detection MCP51 IDE loading adm74xx which would be the first driver detected. I figured that the nolapic was a typo ;-) Hope this helps. At this point if you need anything else with this bug let me know. As far as I'm concerned, you've pinpointed the issue and can 'resolve' this x86_64 bug. Thanks
I expect this is a duplicate of bug #232013. Can you follow up this one and confirm whether it's a duplicate, pls. E.g. try the params that work for those guys and/or try the 2.6.20 kernel pointed to there.
Please mark this has a duplicate of 232013 and continue thread from there. I read in the kernel 2.6.20 changelog that there was an apic issue with commit fea5f1e19611d94fbf3905875a427c4cb959cd06 which was reverted. I suspect it relates to issues pointed out with bug 232013 for some kernel rcs and should be fixed since that revert. I'll monitor the thread of bug 232013 from now on. I'm curious though, as was requested in that bug, to know if a new 10.2 master with updated kernel will be release to fix a few of those 'show stopper' bugs... Thanks
I doubt it has to do with the ATI chipset detection and double as fast running timer issue. Both could be a duplicate of 231205 where someone detected the patch: patches.arch/i386-apic-auto as the culprit. It's dangerous to modify blacklisting code... but as so many people seem to be affected, I try to find something ... *** This bug has been marked as a duplicate of bug 231205 ***